Comparative data for DHA incorporation into NDoPE and conversion to synaptamide by Neuro2A cultures during a 24 h incubation is shown in Figure 1. The cells were incubated with a mixture containing 2 μM unesterified [¹³C]DHA and 2 μM DHA-lysoPC containing 87% sn-1-DHA-lysoPC, in the presence of the fatty acid amide hydrolase (FAAH) inhibitor URB597. A linear increase in [¹³C]DHA incorporation into N-docosahexaenoyl moiety of NDoPE occurred during the first 8 h, but no further appreciable increase was observed thereafter. Although the time-dependent increase of NDoPE production from DHA-lysoPC continued over the entire 24 h incubation, unesterified [¹³C]DHA produced 1.3 to 5.8 times more NDoPE than DHA-lysoPC (Figure 1A).

Mass spectrometric analysis of the NDoPE demonstrated that 75 to 90% of the total DHA incorporated from both substrates was present in plasmalogen-NDoPE (pNDoPE) (Figure 1B) in comparison with diacyl-NDoPE (Figure 1C). For example, at 8 h, 4.6 times more DHA derived from unesterified DHA was incorporated into pNDoPE than into NDoPE. Further analysis of the pNDoPE indicated that more DHA from both DHA substrates was incorporated into the 16:0-plasmalogen NDoPE species (p16:0-NDoPE) than either p18-NDoPE or p18:1-NDoPE at each time point (Figure 1D).

Synaptamide production was measured in the incubation medium [2,3,28]. The amount produced by the Neuro2A cells from unesterified [¹³C]DHA increased linearly during the first 8 h of incubation and then more slowly between 8 and 24 h, whereas synaptamide production from DHA-lysoPC steadily increased throughout the 24 h incubation (Figure 1E). Between 4 to 24 h, 1.7 to 4.2 times more synaptamide was synthesized from unesterified [¹³C]DHA than from DHA-lysoPC.

Only a minute fraction of the DHA incorporated into NDoPE was converted to synaptamide in Neuro 2A cells. After 4 h, the medium contained 26 fmol synaptamide produced from unesterified [¹³C]DHA and 8 fmol produced from DHA-lysoPC. Approximately 3400 fmol of NDoPE was produced from unesterified [¹³C]DHA and 600 fmol from DHA-lysoPC, indicating that synaptamide production accounted for only 0.76–1.3% of the DHA incorporated into total NDoPE. A similar result was obtained after 8 h; synaptamide production accounted for only 1.2–2.7% of the unesterified DHA incorporated into NDoPE and 0.80–0.91% of the DHA from DHA-lysoPC incorporated into NDoPE. Even after 24 h of incubation, synaptamide production accounted for only 1.4–3.3% and 0.80–2.0% of the DHA from unesterified DHA and DHA-lysoPC incorporated into NDoPE, respectively.

Figure 2 shows the concentration dependence of DHA incorporation into total NDoPE, pNDoPE, diacyl NDoPE, and synaptamide. The Neuro2A cells were incubated separately with increasing amounts of [¹³C]DHA or [¹³C]DHA-lysoPC for 8 h at 37 °C. At each substrate concentration, substantially more DHA from unesterified [¹³C]DHA than [¹³C]DHA-lysoPC was incorporated into total NDoPE (Figure 2A), pNDoPE (Figure 2B), and diacyl NDoPE (Figure 2C). At each substrate concentration, pNDoPE accounted for about 80% of the total NDoPE derived from unesterified [¹³C]DHA or [¹³C]DHA-lysoPC. Among pNDoPE species, more DHA from each substrate was incorporated into the p16:0-NDoPE than either p18:0-NDoPE or p18:1-NDoPE (Figure 2D). Increasing amounts of synaptamide were also produced as the concentration of unesterified [¹³C]DHA or [¹³C]DHA-lysoPC was raised (Figure 2E), but from 1.5 to 2.4 times more synaptamide was produced from unesterified [¹³C]DHA. Consistent with the results in Figure 1, the amount of synaptamide produced was only 0.67–1.1% of the DHA incorporated into total NDoPE in the incubations with unesterified [¹³C]DHA, and 0.58–1.8% in the incubations with [¹³C] DHA-lysoPC.

Figure 3 compares the total uptake of DHA in 8 h by Neuro2A cells incubated with increasing amounts of either unesterified DHA or DHA-lysoPC, as measured by gas–liquid chromatography (GC). Increasing amounts of DHA were taken up from both substrates over this range of concentrations. However, at 2, 4, and 8 μM, the DHA uptake from unesterified DHA was 55, 65, and 90% greater than the amount taken up from DHA-lysoPC.

On the basis of these comparative uptake results, we recalculated production of NDoPEs and synaptamide from unesterified DHA and DHA-lysoPC to account for the differential cellular incorporation of DHA from these precursors. Figure 4 shows that the production of total NDoPE (Figure 4A), pNDoPE (Figure 4B), diacyl NDoPE (Figure 4C), NDoPE molecular species (Figure 4D), and synaptamide (Figure 4E) from unesterified DHA and DHA-lysoPC was similar when normalized to account for the differences in DHA uptake. Statistical analysis after normalization indicated that only NDoPE production at 1 and 2 μM showed significant differences between these two substrates, while no significance was found for synaptamide production. Normalization did not alter the higher production of pNDoPE in comparison diacyl-NDoPE, and p16:0-NDoPE still remained the most abundant NDoPE species.

The cytotoxicity of N-acylphosphatidylethanolamine-phospholipase D (NAPE-PLD) inhibition was assessed in Neuro2A cultures. Previous studies indicated that hexachlorophene and bithionol produced only small decreases in HEK293 cell viability at concentrations of 8 and 10.5 μM, respectively [29]. On the basis of this information, we tested the effect of these inhibitors on Neuro2A cells at a concentration of 10 μM. As shown in Figure 5, no cytotoxicity was observed in 1.5 h incubations, but a 10–20% reduction in cell viability occurred at 2.5 h. Therefore, a 2 h incubation time, sufficient to obtain reproducible amounts of synaptamide production (Figure 1), was utilized to test the effect of the inhibitors on synaptamide production.

As seen in Figure 6, 10 μM hexachlorophene inhibited the production of [¹³C]synaptamide by 65% in Neuro2A cultures incubated with either 2 or 4 μM [¹³C]DHA in the presence of URB597 (Figure 6A, middle set of bars). Bithionol inhibited [¹³C]synaptamide production to a similar extent (Figure 6B, middle set of bars). The production of synaptamide (left set of bars) and anandamide (right set of bars) derived from endogenous substrates was also reduced by these NAPE-PLD inhibitors.

Uniformly [¹³C]-labeled DHA ([¹³C]DHA) was a gift from Dr. Anthony Windust (National Research Council, Ottawa, Ontario, Canada), and we synthesized unlabeled and uniformly [¹³C]-labeled DHA-lysoPC. Unlabeled DHA-lysoPC was made by lipolysis of 1-palmitoyl,2-docosahexaenoyl-glycerophosphorylcholine (16:0,22:6-PC) from Avanti Polar Lipids (Alabaster, AL, USA). Briefly, 16:0,22:6-PC in ethylacetate–water (50:50) was incubated with the immobilized lipozyme (Novo Nordisk, Bagsvaerd, Denmark) in the dark at room temperature under nitrogen for 24 h. After centrifugation of the immobilized enzyme, and supernatant concentration, DHA-lysoPC was separated from the remaining 16:0,22:6-PC by thin-layer chromatography with chloroform–methanol–water 60:30:4 as eluant. DHA-lysoPC was quantified on the basis of DHA-methylester obtained by transmethylation, measured by capillary GC. [¹³C]-labeled DHA-lysoPC was obtained by acylation of glycerophosphorylcholine (GPC) according to a method described to prepare 1-acetyl,2-DHA-GPC [45] without the acetylation step. GPC was dissolved in dimethylformamide, and added to 1,3- dicyclohexylcarbodiimide, 4-dimethylamino-pyridine, and [¹³C]DHA for incubation in the dark at room temperature under nitrogen for 24 h. The resulting [¹³C]DHA-lysoPC was purified by thin-layer chromatography and measured according to its DHA content by GC, as stated above for unlabeled DHA-lysoPC.

Diacyl and plasmalogen N-17:1 (number of carbons/number of double bonds)-NAPE were synthesized and purified as internal standards for NDoPE on the basis of procedures described in the literature [46]. Briefly, for diacyl species, 18.8 μL (13.3 mM) 15:0,d7-18:1-PE (Avanti Polar Lipids, Alabaster, AL) in dichloromethane (DCM; Sigma-Aldrich, St. Louis, MO, USA) was added to 1.5 mL DCM. A total of 25 μL of pyridine (Sigma-Aldrich) diluted in DCM (1:25 ratio) was added, followed by 62.5 μL 100 mM 4-dimethylaminopyridine (Sigma-Aldrich) and 65 μL 100 mM 17:1 acid chloride (NuChek Prep, Elysian, MN, USA), both in DCM. The mixture was vortexed and incubated overnight at room temperature. The reaction was quenched with saturated NH₄CO₃ (Sigma-Aldrich) and centrifuged for 3 min at 2500 rpm (1294 RCF), and the upper phase was discarded. The organic layer was further washed with water several times, dried under N₂, suspended in 1 mL methyl-t-butyl ether (MTBE; Sigma-Aldrich)/chloroform (CHCl₃; Thermo Fisher Scientific, Pittsburgh, PA, USA)/acetic acid (Thermo Fisher Scientific) (98:2:0.2), and subjected to solid phase extraction (SPE) (Extract-Clean silica, 1000 mg/1.5 mL, American Chromatography Supplies, Vineland, NJ, USA), similar to the method described in [47]. N-17:1-PE was eluted into MTBE/CHCl₃/methanol (Thermo Fisher Scientific) (50:20:30). The eluant was dried under argon and stored at −80 °C in 2:1 CHCl₃/methanol containing 50 mg/L t-butyl hydroxytoluene (BHT, Sigma-Aldrich). For the synthesis of plasmalogen N-17:1-PE, p18:0,20:4-PE (Avanti Polar Lipids) was utilized as substrate.

Deuterated synaptamide internal standard (d4-synaptamide) was synthesized by dropwise addition of 100 μL deuterated ethanolamine (d4-ethanolamine, Cambridge Isotope Laboratories, Tewksbury, MA, USA) into ice-cold DCM solution containing 100 mg docosahexaenoyl chloride (NuChek Prep). The mixture was incubated for 15 min on ice and washed several times with liquid chromatography/mass spectrometry (LC–MS)-grade water (Thermo Fisher Scientific) until the final 2 washes were pH-neutral. The organic layer was dried under N₂, suspended in methanol, and stored under argon at −80 °C.

The quantity of lipid substrates, N-17:1-PE, and d4-synaptamide internal standards was calibrated by GC after transmethylation. The positional (1- versus 2-DHA-lysoPC) as well as ¹³C isotopic distribution of DHA-lysoPC and unesterified DHA was characterized by LC–high-resolution MS in both positive and negative mode, using LC conditions described for the synaptamide analysis (Figure 7). Approximately 87% of DHA-lysoPC was 1-DHA-lysoPC (Figure 7A), and 68% of [¹³C]DHA was the fully labeled isotope, [¹³C₂₂]DHA (Figure 7B).

Mouse neuroblastoma Neuro2A cells (ATCC, Manassas, VA, USA) were grown in glutamine-free, low glucose Dulbecco’s minimum essential medium (DMEM; Sigma-Aldrich) containing 5% fetal bovine serum (FBS; Sigma-Aldrich) and penicillin/streptomycin antibiotic (Gibco, Gaithersburg, MD, USA) at 37 °C in a 5% CO₂ atmosphere for up to 4 passages. Prior to treatment with lipid substrate(s), approximately 2.5 × 10⁴ cells/cm² were seeded into 6-well plates or 10 cm cell culture dishes, and after 24 h, the cultures were transferred to 0.25% FBS media. A FAAH inhibitor URB597 (Sigma-Aldrich) was included to prevent synaptamide hydrolysis [2,3,28].

The lipid substrate for the time-dependent studies was a mixture of unlabeled DHA-lysoPC and [¹³C]DHA, while the concentration-dependent studies used each [¹³C]-labeled substrate separately. Dimethyl sulfoxide (DMSO; Sigma-Aldrich) stocks of the URB597 fatty acid amide hydrolase (FAAH) inhibitor and vitamin E (Sigma-Aldrich) containing appropriate lipid substrates were prepared. At 24 h after transfer to the 0.25% FBS medium, DMSO mixture was added to 0.25% FBS in DMEM, such that final media contained 0.1% DMSO, 2 μM URB597, 40 μM vitamin E, and 2 μM of each lipid substrate for time-dependence studies. For the concentration-dependence studies, each labeled substrate at 1–8 μM was incubated for 8 h. At the end of incubation, the media was collected and BHT-methanol was added to make a 30% aqueous solution. Cells were washed twice with phosphate-buffered saline (PBS; Gibco). Both media and cells plates, with 0.8 mL PBS added per well, were stored at −80 °C until processing.

Cells were scraped into glass tubes and suspended in PBS/BHT-methanol/chloroform (0.8:2:1) with internal standards including a diacyl and plasmalogen N-17:1-NAPE (15:0,d7-18:1 and p18:0,20:4 species), deuterated phospholipid species (d35-18:0,18:1-PE, p18:0,d9-18:1-PE, d35-18:0-lysoPC (Avanti Polar Lipids)), and 23:0 (Nu-Chek Prep). Lipids were extracted according to the method of Bligh and Dyer [48]. Depletion of phospholipids from 90% of the Bligh–Dyer extracts was achieved by silica SPE using the same solvent system described for N-acyl 17:1 NAPE synthesis [38]. The SPE eluant was dried and chemically deacylated in 240 μL of methylamine (40% aqueous, Sigma-Aldrich)/BHT-methanol/1-butanol (Sigma-Aldrich) (4:4:1) mixture at 53 °C [39] for 90 min. The product was dried under N₂, resuspended in 1 mL of BHT-methanol/water containing 0.1% acetic acid (Thermo Fisher Scientific) (7:3), and loaded onto strata-X polymeric reverse-phase SPE cartridges (Phenomenex, Torrance, CA, USA) pre-equilibrated with water. After washing with 1 mL of the loading solvent followed by 3 mL water (neutral), deacylated NDoPE species were eluted with 3 mL of BHT-methanol, dried under N₂, and suspended in approximately 17 μL of BHT-methanol for LC–MS/MS analysis (Figure 8A).

Deacylated NDoPE samples were injected onto a Gemini-NX C18 LC column (150 mm × 2.1 mm, 5 μ, Phenomenex), which was coupled to Q-Exactive mass spectrometer to detect the four resulting classes of deacylated compounds, N-docosahexaenoyl-glycerylphosphatidylethanolamine (NDoGPE), 1-p16:0-N-DHA-GPE, 1-p18:0-NDoGPE, 1-p18:1-NDoGPE, and 1,2-lyso-N-DHA-GPE. LC separation was achieved utilizing a 2-solvent system at 0.4 mL/min flow rate, as described previously [49]: 0.1% NH₄OH (Fisher Scientific) in solvent A (88%/12% methanol/water) and B (88%/12% methanol/n-hexane (Fisher Scientific)). The gradient consisted of 100% A for 3 min, followed by a linear change to 100% B over 18 min. The column was held at 100% B for an additional 6 min. Between each sample, the column was reconditioned with 0.1 M ammonium acetate (Sigma-Aldrich) for 3 min followed by 100% A for 3 min. The mass spectrometer was run in positive MS/MS mode, and quantitation of each deacylated NDoPE class was achieved via the fragment ions [39] arising from transitions 526.3 to 354.279 (NDoGPE), 548.4 to 376.353 ([13C]NDoGPE), 748.5 to 354.279 (1-p16:0-NDoGPE), 770.6 to 376.353 (1-p16:0-[13C]NDoGPE), 776.6 to 354.279 (1-p18:0-NDoGPE), 798.6 to 376.353 (1-p18:0-[13C]NDoGPE), 774.5 to 354.279 (1-p18:1-NDoGPE), and 796.6 to 376.353 (1-p18:1-[13C]NDoGPE). Peak areas of these transition fragments were compared to those from the N-17:1-PE internal standards: 466.3 to 294.279 (N-17:1-GPE) and 716.6 to 294.279 (1-p18:0-N-17:1-GPE) for diacyl and plasmalogen species, respectively.

The procedure for synaptamide analysis is illustrated in Figure 8B. A mixture of deuterated internal standards including d₄-synaptamide, d₅-DHA, and d₈-AA (deuterated free fatty acids, Cayman Chemical, Ann Arbor, MI, USA) was added to media samples, which were made 7:3 BHT-methanol/water and centrifuged for 20 min at 4 °C. Supernatants were loaded onto Strata-X polymeric C18 reverse-phase SPE cartridges (33 μ, 30 mg/mL, Phenomenex) that were wetted with BHT-methanol and equilibrated with water. After being washed with water, samples were eluted with 2.5 mL BHT-methanol into glass tubes, dried under N₂, and suspended in 25 μL BHT-methanol. An Eclipse C18 UHPLC column (1.8 μ, 2.1 mm × 50 mm, Agilent Technologies, Santa Clara, CA, USA) was coupled to a high-resolution Thermo Scientific Q-Exactive mass spectrometer for analysis. A 3-part tertiary gradient, with all solvents containing 0.01% acetic acid, consisting of water (A), methanol (B), and acetonitrile (Avantor, Radnor Township, PA, USA) (C) was used for LC. After pre-equilibration of column with A/B (60%/40%), 5 μL extract was injected and the solvent composition was linearly changed to A/B/C (36.3%/15%/48.7%) in 5 min, followed by a linear gradient to A/B/C (13.5%/68.4%/18.1%) over 22 min. Positive ion MS/MS was utilized to detect natural, [¹³C]-, and d₄-synaptamide, using mass transitions of 372.3 to 62.060, 394.4 to 62.060, and 376.3 to 66.085, respectively. The fully labeled [¹³C]synaptamide signal (394.4 to 62.060) was corrected according to the unique isotope distribution for the synaptamide production from the ¹³C-labeled substrates. Quantitation of fatty acid and DHA-lysoPC was achieved by comparison of peak areas of [M + Acetate]⁻ or [M + H]⁺ ions to that of the corresponding internal standard.

The remaining 10% of Neuro2A cell Bligh–Dyer lipid extract (see deacylated NDoPE analysis) was transmethylated with boron trifluoride/methanol (Sigma-Aldrich) in the presence of a 23:0 fatty acid internal standard NuChek Prep), and the reaction product was extracted into hexane (Thermo Fisher Scientific). GC analysis was similar to that previously described [49]. The transmethylated samples were injected onto an Agilent 6890 gas chromatograph with a flame ionization detector via a 15 m DB-FFAP phase capillary column (Agilent Technologies). Fatty acid methyl esters were identified according to the elution time and quantified on the basis of the peak area in comparison to the 23:0 internal standard.

Effects of the NAPE-PLD inhibitors hexachlorophene and bithionol (Sigma-Aldrich) [29] on cell viability were tested in the presence of 2 μM FAAH inhibitor URB597. Neuro2A cells were seeded at approximately 40% confluency into a 96-well culture plate in 5% FBS/DMEM medium. Then, 24 h later, the media was aspirated and replaced with 0.25% FBS/DMEM containing 2 μM URB597 and 0, 5, or 10 μM hexachlorophene or bithionol, and the incubation was continued for 1.5 or 2.5 h at 37 °C. Cells were lysed and viability determined by an ATP-based luminescence assay (CellTiter-Glo Luminescent Cell Viability Assay Kit, Promega, Madison, WI).

The effect of these NAPE-PLD inhibitors on [¹³C]synaptamide production was also tested in the presence of 2 μM URB597. Neuro2A cells were seeded into 6-well plates (5% FBS/DMEM) and incubated overnight. After changing the medium to 0.25% FBS/DMEM containing 2 μM URB597 and 10 μM hexachlorophene or bithionol and incubating for 15 min at 37 °C, we added [¹³C]DHA to each well along with 40 μM vitamin E, and incubated it for 2 h at 37 °C. Media was collected and d4-synaptamide internal standard was added. Media samples were processed and analyzed as described above. For this assay, endogenous anandamide was also quantitated, utilizing d₄-synaptamide as an internal standard.

The quantitative results are expressed as means ± standard deviation for triplicate samples, except for viability test of Neuro2A cells incubated with NAPE-PLD inhibitors, where means were based on quadruplicate samples. Statistical analyses were conducted using Student’s t-test (Excel software, Microsoft, Redmond, WA, USA). Statistical significance is reported at * p < 0.05, ** p < 0.01, and *** p < 0.001.